INDEX

Explanations

references to specific entities, particularly related to classification or categorization

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Text related to dedication or allocation

np_acts-logits-general · gemini-2.0-flash

dedicated or devoted to

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 oprot

-0.71

AndEndTag

-0.65

IntoConstraints

-0.59

 φο

-0.57

npmjs

-0.55

oredCriteria

-0.55

politan

-0.55

ChildScrollView

-0.54

 rootReducer

-0.54

albero

-0.54

POSITIVE LOGITS

 dedicated

2.23

 reserved

2.11

 devoted

1.93

dedicated

1.90

 dedicate

1.89

Dedicated

1.76

 Dedicated

1.76

 dedic

1.72

 allocated

1.69

reserved

1.64

Activations Density 0.300%

references to specific entities, particularly related to classification or categorization

Text related to dedication or allocation

dedicated or devoted to

No Comments

No Known Activations